Goodness-of-Fit Measures Data analysis

22 Average consumption in the country during the years 2000-2010 amounted Libya 21,211,206 MWh, consumption 14,522,025 MWh lowest and highest value of the consumption at 31,680,704 MWh. The total amount of consumption in 2000-2010 amounted to 233,323,266 MWh.

4.2.1.4 Statistic descriptive of population

Population from year 2000-2010 in the State of the lowest Libya reach 5,231,189 person and 6,355,112 person achieve complete biggest population on Libyan state as the below table. Table 4.6 Description of Population from 2000-2010 N Valid 11 Missing Mean 5,785,868 Median 5,769,709 Mode 5,231,189 a Std. Deviation 383,063 Minimum 5,231,189 Maximum 6,355,112 Sum 63,644,551 Population in the country of Libya reached 5,785,868 person. the entire population of the year 2000-2010 reached number 63,644,551 person

4.2.2 Goodness-of-Fit Measures

Table of summary statistics for stationary R-square, R-square, root mean square error, mean absolute percentage error, mean absolute error, maximum absolute percentage error, maximum absolute error, and normalized Bayesian Information Criterion. The Normalized Bayesian Information Criterion, as a general measure of the overall fit of a model that attempts to account for model complexity. It is a score based upon the mean square error and includes a penalty for the number of parameters in the model and the length of the series. The penalty removes the advantage of models with more parameters, making the statistic easy to compare across different models for the same series Norušis, 2007. commit to user 23 Table 4.7 Model Fit Statistic Model Stationary R- squared R-squared RMSE MAPE Production-Model_1 .580 .991 884,1431.741 1.763 Generated-Model_2 .580 .991 57,4491.991 1.763 Consumption- Model_3 -1.757E-16 .958 1,217,524.605 3.815 Population-Model_4 0.000 1.000 4,947.405 .071 Continue Table 4.7 Model MAE MaxAPE MaxAE Normalized BIC Production-Model_1 6,021,407.853 4.308 19,004,280.541 32.426 Generated-Model_2 391,254.571 4.308 1,234,846.040 26.958 Consumption- Model_3 783,931.280 10.868 3,015,625.100 28.255 Population-Model_4 4,272.438 .125 7,839.250 17.273 Stationary R - squared. A measure that compares the stationary part of the model to a simple average. Variables that have a negative value that is consumption, which means that the model under consideration is worse than the base line model. While the variable production costs, generated, population have a positive value means that the model under consideration is better than the basic model. R - Squared. Estimates of the proportion of the total variation in the series described by the model. This step is very useful when the series is stationary. All variables have a positive value means that the model under consideration is better than the basic model. RMSE. Root Mean Square Error. The square root of the mean square error. A measure of how much the series varies depending on the level of its prediction models, expressed in the same units as the dependent series. RMSE of all the variables with the highest consumption variable 1,217,524.605 value and a low of population variable with a value of 4,947.405 MAPE. Absolute Percentage Error means. A measure of how much the series varies depending on the level of its prediction models. It does not depend on the units used and can therefore be used to compare series with different units. commit to user 24 As for who has the highest MAPE lowest in the population variable with a value of 0.071 MAE. Mean absolute error. Measures how much the series varies from its level prediction models. MAE is the highest on the production variable value and the lowest 6,021,407.853 population variable with a value of 4,272.438 MaxAPE. Maximum Absolute Percentage Error. The largest estimated error, expressed as a percentage. MaxAPE highest in the variable value consumption by 10.868 and the population with the lowest value of 0.125. MaxAE. Maximum Absolute Error. The largest estimated error. MaxAE highest in variable production abaout 19,004,280.541 and the lowest values occurring in the population variable with a value of 7,839,250. Normalized BIC. Normalized Bayesian Information Criterion. General measure overall model fit that tries to explain the complexity of the model. Normalized BIC variables that have high production is a variable with a value of 32,426 and the lowest values occur in the population variable with a value of 17. 273.

4.2.3 Result Analysis ARIMA Model Parameters by SPSS